Fish Meal Project
-
Upload
akicontractingitaldeco -
Category
Documents
-
view
226 -
download
1
Transcript of Fish Meal Project
-
7/25/2019 Fish Meal Project
1/87
Electrically powered drying of fish meal
Magns Valgeir Gslason
Faculty of Industrial Engineering,
Mechanical Engineering andComputer Science
University of Iceland2013
-
7/25/2019 Fish Meal Project
2/87
-
7/25/2019 Fish Meal Project
3/87
Electrically powered drying of fish meal
Magns Valgeir Gslason
60 ECTS thesis submitted in partial fulfillment of aMagister Scientiarumdegree in Mechanical Engineering
AdvisorsSigurjn Arason MSc, Professor, University of Iceland, MatsDr. Halldr Plsson, Associate Professor, University of Iceland
Dr. Bjrn Margeirsson, Research Group Leader, Mats
Faculty RepresentativeDr. lafur Ptur Plsson, University of Iceland
Faculty of Industrial Engineering, Mechanical Engineering and ComputerScience
School of Engineering and Natural SciencesUniversity of IcelandReykjavik, May 2013
-
7/25/2019 Fish Meal Project
4/87
Electrically powered drying of fish meal60 ECTS thesis submitted in partial fulfillment of aMagister Scientiarumdegree inMechanical Engineering
Copyright 2013 Magns Valgeir GslasonAll rights reserved
Faculty of Industrial Engineering, Mechanical Engineering and Computer ScienceSchool of Engineering and Natural SciencesUniversity of IcelandHjararhagi 2 - 6107, ReykjavikIceland
Telephone: 525 4000
Bibliographic information:Magns Valgeir Gslason, 2013,Electrically powereddrying of fish meal, Masters thesis,Faculty of Industrial Engineering, Mechanical Engineering and Computer Science,University of Iceland, pp. 87.
Printing: Hsklaprent, Flkagata 1, 107 ReykjavikReykjavik, Iceland, May 2013
-
7/25/2019 Fish Meal Project
5/87
AbstractThe fish meal industry is an important sector within the Icelandic fish industry and has
applied technology in recent years. Fish meal processing is an energy intensive process.
For better control of energy utilization in the process a mass- and energy flow model was
set up for processing different raw material, and simultaneously for a better overview for
the process. The model is a good tool to influence the quality of the fish meal products.
The main aim of the project was to analyse the energy usage and contribute to improved
energy utilization specially for the air drying with the new electric air heating equipment.
The drying is the last step in the process and waste heat is utilized at previous stages in the
process.
The aim of the project is to estimate the benefits of using electricity to heat air for drying
of fish meal. In that way it would be possible to reach the goal for the Icelandic marine
sector to utilize exclusively domestic renewable energy for fish meal processing, reduce
imports of oil for shore processing and reduce the carbon footprint. Measurements in the
process were carried out for four kinds of raw material for evaluation of mass flow through
the process.
The electric air heater has proved its worth in HB Grandi fish meal factory in
Vopnafjordur, Iceland in terms of operational security, energy utilization, control and
maintenance.
-
7/25/2019 Fish Meal Project
6/87
-
7/25/2019 Fish Meal Project
7/87
v
tdrtturFiskmjlsinaurinn er mikilvg atvinnugrein og hefur tknivst miki sustu rum.
Mikil orka er notu vi fiskmjlsvinnslu. Til ess a n betri yfirsni yfir orkunotkun
ferlinum var sett upp massa- og orkustreymislkan fyrir vinnslu mismunandi hrefni sem
hefur fr me sr betri yfirsn yfir vinnslursina. Lkani stular einnig a v a
auveldara er a hafa hrif gi fiskmjlsafura me ferlastringu. Megin markmi me
verkefninu er a greina orkunotkun vinnsluferlinum og srstaklega vi urrkun og
skoa auvelda innleiingu notkunar rafhitunar-bnaar fyrir lofturrkun fiskmjls.
urrkunin er sasta stigi vinnslursinni og kanna notkun glatvarma fr urrkun sem
san er notaur framar rsinni.
Markmi verkefnisins er a kanna mguleika ess a nta rafmagn til a hita loft fyrir
urrkun fiskmjli hagkvman htt. Me v mti vri mgulegt a n v markmii
sjvartvegsins a nta eingngu innlenda orku vi framleislu fiskmjls, draga verulega
r innflutningi olu til landvinnslu og draga tluvert r myndun stspors. Mlingar
framleisluferli voru framkvmdar fyrir fjrar gerir af hrefnis, til a meta efnisstrauma
gegnum verksmijuna. rstifall yfir olukyndingarbna var mlt og er mun meira
samanbori vi rafhitunarbna.
Rafhitunarbnaurinn hefur reynst vel fiskmjlsverksmiju HB Granda Vopnafiri hva
varar rekstrarryggi, orkuntingu, stringu og vihald.
-
7/25/2019 Fish Meal Project
8/87
-
7/25/2019 Fish Meal Project
9/87
vii
Table of ContentsList of Figures ..................................................................................................................... ix
List of Tables ....................................................................................................................... xi
Abbreviations ..................................................................................................................... xii
Acknowledgements ........................................................................................................... xiii
1 Introduction ..................................................................................................................... 1
2 Background ..................................................................................................................... 5
2.1 Fish meal factories in Iceland .................................................................................. 52.2 Raw material............................................................................................................ 8
2.2.1 March capelin whole and cut offs .................................................................. 82.2.2 Mackerel cut offs ........................................................................................... 82.2.3 A mix of herring and mackerel cut offs ......................................................... 82.2.4 February capelin............................................................................................. 9
2.3 Drying ...................................................................................................................... 92.4 Fish meal process .................................................................................................... 9
2.4.1 Material flow ................................................................................................ 102.4.2 Two steps drying of fish meal ...................................................................... 15
2.5 DynoJet indirect air dryer .................................................................................. 172.5.1 Electric heating equipment .......................................................................... 182.5.2 Air flow in Dyno-Jet and air dryers ............................................................. 202.5.3 Steam dryingAir drying ........................................................................... 22
2.6 Energy balance ...................................................................................................... 242.7 Material and energy balances ................................................................................ 25
2.7.1 Mass balance ................................................................................................ 252.7.2 Energy balance ............................................................................................. 25
3 Materials and methods ................................................................................................. 273.1 Mass flow modelling ............................................................................................. 273.2 Air flow measurements for Dyno-Jet .................................................................... 313.3 Pressure drop measurement ................................................................................... 323.4 Energy calculations and comparison ..................................................................... 32
4 Results ............................................................................................................................ 35
4.1 Mass and energy flow model for fish meal process .............................................. 354.1.1 March capelin whole and cut offs ................................................................ 394.1.2 A mix of herring and mackerel cut-offs ....................................................... 394.1.3 Mackerel cut-offs ......................................................................................... 404.1.4 February capelin whole ................................................................................ 41
4.2 Energy balance for evaporation and drying .......................................................... 45
4.2.1
Comparison of waste heat evaporator capacity between raw materials ...... 45
-
7/25/2019 Fish Meal Project
10/87
viii
4.3 Energy utilized for processing fish meal and oil. ................................................... 464.4 Air flow and pressure drop in heat exchangers ...................................................... 50
4.4.1 Air flow for the electrically powered DynoJet air dryermeasurement ................................................................................................ 50
4.4.2 Pressure drop over oil heating equipment ................................................... 51
5 Discussion ....................................................................................................................... 53
6 Conclusion ...................................................................................................................... 59
7 Bibliography ................................................................................................................... 61
Appendix A ......................................................................................................................... 67
Appendix BPressure drop measurements .................................................................... 69
Appendix C ......................................................................................................................... 71
-
7/25/2019 Fish Meal Project
11/87
ix
List of FiguresFigure 1: Evolution of fish meal and crude oil price since 1982 (Index Mundi, 2012). ....... 2
Figure 2: Disposition of landings in Iceland 000 mt (metric tons) (Andersen, 2011). ........ 3
Figure 3: Locations of Icelandic fish meal factories. ............................................................ 6
Figure 4: Fish meal and oil usage estimation for 2010 (Shepherd & Jackson, 2012). .......... 7
Figure 5: Fish meal production in product weight from 2000 and prediction until2021 (FAO, 2012). .............................................................................................. 7
Figure 6: Diagram of the material flow in Vopnafjordur plant. .......................................... 10
Figure 7: Schematic view of MVR evaporator. ................................................................... 13
Figure 8: Schematic view of WHE II evaporator. ............................................................... 14
Figure 9: Viscosity of stick-water in terms of dry matter (Heggernes, 1991). .................... 15
Figure 10: Steam dryer (Arason, 2009). ............................................................................. 16
Figure 11: Dynojet air drying system (Stord, 1992). ....................................................... 17
Figure 12: Installation of the electrical heating equipment in Vopnafjordur. ..................... 19
Figure 13: DynoJet air drying drum (Stord, 1992). ......................................................... 19
Figure 14: Hetland drying drum (Arason, 2009). ................................................................ 21
Figure 15: Drying efficiency of steam dryer in terms of water content of fish meal(Arason, 2013). ................................................................................................. 23
Figure 16: Schematic view of energy balance in the fish meal process. ............................. 24
Figure 17: Measurement points of HB Grandi fish meal factory in Vopnafjordur. ............ 28
Figure 18: Oil heating heat exchanger and pressure drop measuring hose connectedat inlet and outlet of the heat exchanger. Hose connection are marked onthe figure, red ring the air outlet of the heat exchanger, yellow ring is themeasuring position and blue ring is the air inlet to the heat exchanger. ........... 32
Figure 19: Relative mass flow for all for raw materials, MC* in vapour WHE II is theevaporation for both MVR and WHE II run in parallel.................................... 37
Figure 20: Results of chemical analysis for February capelin set up in a process
diagram. ............................................................................................................ 38
-
7/25/2019 Fish Meal Project
12/87
x
Figure 21: Results of air flow measurements. ..................................................................... 50
Figure 22: Where the air enters the electric heater, first step is steam tube and fin heatexchanger (Sigmundsson, 2011). ..................................................................... 57
Figure 23: Same view as in Figure 22 after one day in operation (Sigmundsson,2011). ................................................................................................................ 58
-
7/25/2019 Fish Meal Project
13/87
xi
List of Tables
Table 1: Capacity of air dryers (Arason, 2013; SR-Mjl, 2013). ....................................... 22
Table 2: Comparison between drying time and meal temperature (Gunnarsson,2003). ................................................................................................................ 22
Table 3: Dates of sampling. ................................................................................................. 27
Table 4: Results of chemical analysis performed by Matis. Abbreviations: MC March capelin, HMCO Herring and mackerel cut offs, MCO
Mackerel cut offs, FCFebruary capelin, M1Measurement 1. ................... 36
Table 5: State of factory when measurement A was carried out. ........................................ 42
Table 6: State of factory when measurement B was carried out. ........................................ 43
Table 7: Evaporation for all raw materials. ......................................................................... 44
Table 8: Waste heat evaporation in terms of waste heat input from steam dryer fordifferent raw materials. ..................................................................................... 45
Table 9: Evaporation in water removal processes. .............................................................. 45
Table 10: Comparison between processing properties mass flow of raw materials. ........... 45
Table 11: Energy consumption comparison between factories (UST, 2012). ..................... 48
Table 12: Setup of factories in Table 11. ............................................................................ 49
Table 13: Energy cost comparison between heavy fuel oil (HFO) and electricity(Hgnason, 2012; Ptursson & Bjrgvinsson, 2013). ...................................... 50
Table 14: Measured pressure drop in oil heating heat exchanger. ...................................... 51
Table 15: Comparison between oil and electricity in terms of efficiency and costs(Hgnason, 2012; Sigmundsson, 2013; Ptursson & Bjrgvinsson, 2013). ..... 54
Table 16: Results of air flow measurements through electric air heater with pitot tubeair speed meter ( PCE - Deutschland GmbH, 2013) on 14th of January2011, the diameter of the air duct is 0.9 m. ...................................................... 67
-
7/25/2019 Fish Meal Project
14/87
xii
AbbreviationsAKN Akranes
D-J DynoJet air dryer
EHE electrical heating equipment
FC February capelin
FES Isfelag Vestmannaeyja
FFDM fat free dry matter
h hour
HBG HB GrandiHFO heavy fuel oil
HMCO herring and mackerel cut offs
ISK Icelandic krona
MCO mackerel cut offs
MC March capelin
mt metric ton
MVR mechanical vapour recompression
MInput mass input into processNOR Nordfjordur
Pa*s Pascal second, unit for viscosity
RM raw material
RInput drying ratio of fat free dry matter input
RFFDM Input ratio of fat free dry matter input
RFFDM Output ratio of fat free dry matter output
SD steam dryer
SEY Seydisfjordur
STAHE steam to air heat exchanger
SVN Sildarvinnslan
VES Vestmannaeyjar
VSV Vinnslustod Vestmannaeyja
WHE II two step waste heat evaporator
TOH Torshofn
-
7/25/2019 Fish Meal Project
15/87
xiii
AcknowledgementsThe author would like to thank for the support and guidance of the advisors, Sigurjn
Arason, Bjrn Margeirsson and Halldr Plsson. In addition, I would like to thank Gunnar
Plsson and his co-workers at Hedinn, Stefn Hgnason El-run and Sveinbjrn
Sigmundsson and his great team in HB Grandi fish meal factory in Vopnafjordur. Also I
would like to thank for the financial support from the AVS R&D Found of Ministry of
Fisheries of Iceland (project no. R10 084-10 Rafurrkun fiskmjli). All my friends and
colleagues at Matis food and biotech R&D. And best thanks to Theodr Blndal CEO of
Varma og Vlaverk and Stefn Ragnar Jnsson for proofreading.
-
7/25/2019 Fish Meal Project
16/87
-
7/25/2019 Fish Meal Project
17/87
1
1 IntroductionFish species and cut-offs that arent utilized for human consumption, are used for fish meal
production. The earliest example of fish meal production in Iceland is from 1910 in
Siglufjordur. At first wood tubes were filled up with herring and steam used directly for
heating the herring. The liquid phase was pressed with filter cloth and the oil was utilized,
meanwhile the fish meal was used as a fertilizer. Since then the fish meal process has
developed, and new production units incorporated into the multi-unit process.
Thoughts of using a renewable energy source for fish meal drying are not new, back in the
1980s in a mechanical engineering bachelors project in University of Iceland. In the
thesis it was concluded that energy costs were significantly higher (66 90 %) for
electricity than fossil fuel (Jnsson, 1981). Nowadays it is a vice versa situation where the
oil prices have increased as shown in Figure 1 (Index Mundi, 2012), while cost of
electricity has dropped due to precarious insecure energy agreements (Jnsdttir, 2011).
Price of fish meal goes hand in hand with other commodities in the world market such as
oil, the price trends can be seen inFigure 1.
-
7/25/2019 Fish Meal Project
18/87
2
Figure 1: Evolution of fish meal and crude oil price since 1982 (Index Mundi, 2012).
-
7/25/2019 Fish Meal Project
19/87
3
The trend for the pelagic fish species for fish meal and fish oil production is towards that a
greater share of the landings goes to processing for human consumption while the share of
direct landings for fish meal production has decreased according toFigure 2 (Andersen,
2011). As shown onFigure 2 only one third of the total catch is landed direct to fish meal
and oil production in 2010 compared to nearly two thirds in 2006. AlthoughFigure 2 only
reaches back to 2006 the trend has been the same for the last decades.
Figure 2: Disposition of landings in Iceland 000 mt (metric tons) (Andersen, 2011).
This success can be attributed to better handling and cooling of the raw material on board
the pelagic fishing vessels. Better handling refers to shortening cooling down time,
enhanced pumping systems onboard and for landing and finally a lower ratio of fish versus
sea water in storage tanks inboard. Improved cooling and temperature distribution in
storage tanks in the fishing vessels is the result of measurements and development of the
old cooling system. Improved cooling of the raw material results in higher quality and less
degradation of the fish, not only in processing for human consumption, but also leads to
better quality fish meal and oil (orvaldsson et.al., 2011).
138
208 159 134
226308
163
246
160103
221
273
549
490
519
385
195
285
0
100
200
300
400
500
600
700
800
900
1000
2006 2007 2008 2009 2010 2011 est.
'000 mt
Direct landings
Trimmings
Human consumption
-
7/25/2019 Fish Meal Project
20/87
4
The two main objectives of this thesis are to: (1) first assess the advantages of
electrificationof Icelandic fish meal drying and to make a model of mass and energy flow
for various types of raw materials. (2) Sceond possibilities of recycling waste heat from the
process will be discussed, and what could have been done better in the design of the
electric heating equipment for the Dyno Jet air dryer in HB Grandi fish meal factory in
Vopnafjordur.
-
7/25/2019 Fish Meal Project
21/87
5
2 Background
2.1Fish meal factories in Iceland
Today 11 fish meal factories are situated around half the coastline as shown inFigure 3.
These factories are well equipped, with most of the production computer controlled. The
number of fish meal factories has decreased during the last decades from 54 in 1970
(rnason, Arason,& Jnsson, 1986) dispersed around the coastline in contrast to today,
where the factories are only situated on the East and South Coast. The reduction can be
traced to candidature of whole fish for fish meal manufacturing, and other utilizingpossibilities for ground fish cut offs such as head and backbone drying for African markets
and viscera freezing for fur animals (Arason, 2009 2010; lafsson & Arason, 1999;
Sigursson, 2003).
The main reason for the location of the factories is the distance to the fishing grounds of
pelagic fish. Most of them are situated around the East Coast of Iceland while two are
situated in Vestmannaeyjar, and other two on the South West Coast of Iceland. One
disadvantage of the electrification of Icelandic fish meal factories is the electricaltransmission system, where for some places the transmission capacity is not sufficient
(Landsnet, 2013).
-
7/25/2019 Fish Meal Project
22/87
6
Figure 3: Locations of Icelandic fish meal factories.
The main purpose of fish meal and fish oil production is to utilize fish species that arenormally not used for human consumption. Fish meal is mostly used as a protein source for
aquaculture and the oil is a source of energy for aquaculture feed. Increase in aquaculture
has resulted in increased demand for fish meal and fish oil (FAO, 2010).
Since 1970 capelin has been the most important raw material for Icelandic fish meal and
fish oil production. Other important species are herring, mackerel, blue whiting and
pearlsides (Arason & Jnasson, 2009).
For the last decades the trend for fish meal usage has been that the greatest share of fish
meal is utilized for aquaculture feed, while the share of poultry and ruminant feed is
reduced. The main difference in fish meal factoring between these two groups are strict
requirements on digestibility of the protein in the fish meal for the aquaculture.
Furthermore, availability of whole pelagic fish for fish meal production has reduced, the
largest part of pelagic species are processed for human consumption so share of cut-offs is
growing for fish meal production as shown inFigure 5 (FAO, 2012; Arason, 2009). Eight
-
7/25/2019 Fish Meal Project
23/87
7
out of eleven fish meal factories in Iceland are situated close to pelagic proceessing plants,
but the three other have production licence for grading and capelin roe extraction (UST,
2012).
Figure 4: Fish meal and oil usage estimation for 2010 (Shepherd & Jackson, 2012).
Figure 5: Fish meal production in product weight from 2000 and prediction until 2021 (FAO, 2012).
73%
20%
5%
2%
Fish meal 2010
Aquacultur
Pig
Chicken
Other71%
3%
24%
2%
Fish oil 2010
Aquafeed
Industrial
Refined Edible
Hardened Edibl
-
7/25/2019 Fish Meal Project
24/87
8
2.2Raw material
The fish meal process is controlled by the properties of the raw material, and the properties
of the raw material varies by species, season, freshness, storage and many other external
conditions. The way the properties of the raw material can control the process, is first of allhow much water the presses can take out of the raw material. If the presses cant press out
enough water, the water must be evaporated either in the evaporation process or drying,
which is much more energy-intensive. Also the viscosity of the stick-water controls the
level of concentration in the evaporators, water that is not removed in evaporators is
removed in dryers and more energy is utilized.
2.2.1 March capelin whole and cut offs
In the last weeks of the capelin (Mallotus villosus) fishing season when the capelin is
spawning, the male is separated from the female and sent straight to fish meal processing.
On the other hand the roes are taken out of the females and the discharge is sent to fish
meal processing. When the capelin is spawning the fat content drops dramatically, because
the capelin stops consuming feed and lives on the energy reserves stored as body fat.
Around Iceland spawning usually begins in late February and finishes around the end of
March (Matis, 2013).
When working with cut-offs the fish meal factory depends on raw material from the
pelagic processing plant, and the amount of material available to fish meal processing
depends on the yield of the frozen products (whole, fillets and flaps).
2.2.2 Mackerel cut offs
Mackerel (Scomber scombrus) is caught during the summer in Icelandic waters between
the middle of June until the end of September (makrlveiar, 2011). The state of themackerel depends on the season, age and origin. Around the middle of the fishing season
the fat content reaches its maximum level and lowers there after (rarinsdttir, Jnsson,
& Arason, 2011).
2.2.3 A mix of herring and mackerel cut offs
Herring (Clupea harengus)is caught from summer until December depending on condition
of the two main subspecies Icelandic summer spawning herring stock and the Norwegian
-
7/25/2019 Fish Meal Project
25/87
9
and Icelandic herring livestock. The herring goes into human consumption processing and
only cut-offs (head, backbone and viscera) are available for fish meal factoring (Arason,
2009 - 2010).
2.2.4
February capelin
Before spawning the capelin reaches its maximum fat content and a significant difference
is between before and when spawning in March (Arason, 2009 - 2010). In February whole
capelins both male and female go straight to fish meal processing. Higher fat content in
February capelin than the capelin caught in March, will results in an increased volume of
fish oil.
2.3Drying
When a wet solid is subjected to thermal drying two processes occur simultaneously.
Process no. 1 is the removal of water vapour from the material surface, this process
depends on the external conditions of temperature, air humidity and flow, area of exposed
surface and pressure.
Process no. 2 is the transport of moisture internally within the solid, which is a function of
the physical nature of the solid, the temperature and its moisture content. In a drying
operation any of these processes may be the limiting factor controlling the rate of drying,
although they both proceed simultaneously throughout the drying cycle (Mujumdar, 2006).
The drying rate is governed by the rate at which the two processes proceed. At first the
drying rate is constant when removal of water vapour from the surface occurs.
2.4Fish meal process
A fish meal process is continuous due to the law of the conservation of mass, which states
that mass can neither be created nor destroyed. The amount of mass entering the process is
the same as exiting from the process, in form of fish meal, fish oil, evaporated water and
losses (leakage, sampling and due to cleansing) (Felder & Rousseau, 2005). The majority
of the production time the factories run in equilibrium and the process is in steady state.
-
7/25/2019 Fish Meal Project
26/87
10
When flow rates and temperatures do not change with time, except possibly for minor
fluctuations around constant mean values, the process is operating at steady state.
Figure 6: Diagram of the material flow in Vopnafjordur plant.
According to Figure 6 the material flow throughout the process will be listed up in the
following subsections.
2.4.1 Material flow
The following text will discuss the material flow for HB Grandi fish meal plant in
Vopnafjordur, which is in general like all the other processes in other fish meal plants in
Iceland. The main difference between factories is the dehydrating part of the process,
where alignment and number of evaporators and dryers vary between factories.
2 3
14 1312
11
10
5
74
1
6
8
Fish meal
Fish
oil
Raw
material
Pre heaterPre
filter
Press
Decanter
CentrifugePurifying
CentrifugeWHE II
Steam dryerAir dryerCooler and mill
Cooker
9
MVR
-
7/25/2019 Fish Meal Project
27/87
11
1 Raw material storage
The raw material is stored in uninsulated cylindrical tanks, where the temperature inside
the tanks is dependent on raw material- and outside temperatures. The mean raw material
temperature is estimated to be 4 C. To estimate mean raw material temperature it isnecessary to take into account the temperature of the raw material from the rendering plant,
and from data about mean outside temperature in Vopnafjordur being below 4 C for the
whole year around (Norwegian Meteorological Institute, 2012).
1-2 Preheating
The raw material is heated from 4 C to around 40 to 50 C, with waste heat from the
process. Preheating is used to save energy in cooking, by utilizing waste heat from other
stages in the process. In the Vopnafjordur plant there are two parallel preheating units, one
for each production line.
2-3 Cooking
The objective of cooking is to coagulate proteins, with raw material uniform temperatures
around 95 C (Nygaard, 2010). Because of the high temperature cell walls will break down
causing oil and bounded water to flow free in the material, making pressing of liquid from
the material much easier. Another important role for the cooker is to pasteurize and stop all
bacterial growth and heating is also to stop enzymes, thus preventing the formation of free
fatty acids and degradation of vitamins. The raw material is cooked in two parallel screw
cookers, driven by steam from a boiler. The cooker is designed as a cylinder having a
steam heated jacket and steam heated rotor, designed as a screw conveyor.
34/5 Pressing
The purpose of pressing is to remove most of the oil and water from the solid material.Pressing is done in two steps, with a pre-strainer before twin screw presses. Pressing
establishes two phases, press cake and press liquor. Pressing is one of the most important
steps in the whole process. Too high water content results from an insufficiently pressed
material, thus dryers need to use more energy for water removal. Normal water content in
press cake varies from 40 to 55 % and oil content from 2 to 6 %, water and fat content
varies based on raw material properties such as fish species, age, handling and cooking.
-
7/25/2019 Fish Meal Project
28/87
12
The press cake (5) is transferred to the mixer before drying while the press liquor (4)
passes to the decanter.
46/7 Decanter
The decanter is fundamentally a large horizontal centrifuge that separates the solid material
from the press liquor coming from the presses and pre-strainer. Before entering the
decanter the press liquor is heated up in a tank up to 9598 C (Matis, 2012). The solid
material is mixed with the press cake prior the dryer. The liquid from the decanter is called
separated press liquor (6), and the solid is called decanter-sludge (7). The separated press
liquor is transferred to buffer tanks prior the centrifuges.
6
8 Centrifuges
The centrifuges establish three phases: stick-water, sludge and oil. The stick-water is
transferred to the evaporators for concentration, the sludge is either sent to the blender in
front of the dryers or before the cooker. The oil is purified in a special purifying centrifuge
before storage.
89/10 MVR + WHE II Evaporator
The stick-water comes from centrifuges and most of the water is evaporated, resulting inincreased dry matter content. Generally the FFDM is raised up to 18 % - 30 %, it varies
between fish species because of variable viscosity in stick-water. If the stick-water is
excessively evaporated it will result in the material being too viscous and stick-water
fouling on heating surface will occur resulting in lower efficiency in the evaporators.
By running MVR (Mechanical Vapour Recompression) and WHE II (Waste Heat
Evaporator double effect) in series washing is minimized because fouling occurs mainly at
the last states of the evaporation (Arason, 2009; HB Granda, 2010). In addition by runningthe stick-water first through the MVR evaporator, energy is saved due to the MVR being
driven on full capacity which is more energy efficient rather than WHE II (Geankoplis,
2003) .
-
7/25/2019 Fish Meal Project
29/87
13
Figure 7: Schematic view of MVR evaporator.
Mechanical Vapour Recompression evaporator is a conventional single-effect evaporator
with a compressor to recompress the condensed vapour. The vapour coming over-head is
sent to centrifugal compressor driven by an electrical motor. This recompressed vapour
condenses at a temperature higher than the boiling point of the stick-water and the
temperature difference drives the evaporation. Vapour is again generated and the cycle is
repeated (Geankoplis, 2003).
Stickwater
Compressor
Concentrate
Steam
separator
Condensed vapour
~ 2 bar
~ 1 bar
Vapour from
recompression
-
7/25/2019 Fish Meal Project
30/87
14
Figure 8: Schematic view of WHE II evaporator.
Backward feed two-effect Waste Heat Evaporator where the stick-water enters the
secondary effect prior the first effect. The first effect is defined as where the primaryvapour enters. The main reason for feeding the stick-water backwards is high viscosity
especially when the water content drops (Arason, 2001b).
Condensed vapour
from steam dryer
Vapour from
step I
I II
Stickwater
Stickwater
Vapour from step II for
pre-heating
Vapour from steam
dryer
Condense from
step 1
Concentrate
-
7/25/2019 Fish Meal Project
31/87
15
Figure 9: Viscosity of stick-water in terms of dry matter (Heggernes, 1991).
The viscosity of the stick-water controls the intensity of the condensation of the stick-
water, and the viscosity varies between species as shown inFigure 9.
2.4.2 Two steps drying of fish meal
11 12 Pre-drying with steam dryer
A few main types of steam dryers are used for fish meal drying nowadays. Generally the
steam is passed through central pipe (rotor) with discs or spirals on to increase surface
area, size of heating surface can be from 60 m up to 730 m for the largest dryers.
Capacity of the dryers is proportional to the heating surface, pressure of the steam, the
water content and meal coverage of heating surface. Pressure of the steam inside the pipe
can vary from 1 to 10 bar. Water content of meal blend entering the dryer can be from 55
% up to 70 %, and as shown in Figure 10 the volume of the meal decreases relative to
water content. Therefore capacity of the dryer reduces due to decreased coverage of
heating surface and also because of reduced heat transfer through drier meal.
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
10 20 30 40 50 60
Viscosity(Pa*s)
Dry matter (%)
Capelin
Mackerel
-
7/25/2019 Fish Meal Project
32/87
16
The main advantages of the steam dryers compared to air dryers are energy usage and
possibilities to recycle waste heat for lower energy state prior in the process e.g. during
preheating, cooking or evaporation. Furthermore, steam dryers are not sensitive to water
content and viscosity of the wet meal blend, due to high content of stick-water in the wet
meal blend (Arason, 2013). These advantages of the steam dryers make them an attractive
option as a pre-dryer prior to air dryer. Most of the steam dryers in Norway are used as
pre-dryers (Nygaard, 2010).
Figure 10: Steam dryer (Arason, 2009).
12-13 Secondary drying in air dryer
After drying in the steam dryer the meal is sent to the air dryer who is described in more
detail in Section 2.5. The reason for pre drying in a steam dryer is to maximize the
digestibility of the proteins in the meal (Arason, 2013). The digestibility of the proteins is
not temperature dependent when the water content is over 40 %, when the water content
drops below 40 % the temperature of the meal and drying time determine the digestibility
(Arason, 2013).
13-14 Cooler and grinder
After drying, the fish meal temperature can be from 75 to 98 C and the meal must be
cooled down before storage. Usually when cooling down the meal the moisture in the meal
drops by 0.52 %. The cooler is based on the same fundamental principle as the air dryer
e.g. mixing up the fish meal with air and reaching an equilibrium temperature before
separating the air stream from the meal. Cold outdoor air is used for cooling down the
-
7/25/2019 Fish Meal Project
33/87
17
meal, instead of the hot air utilized in the dryer. The meal is cooled down to around 20 C
with outdoor air with a temperature average ranging from -2 C up to around 12 C
(Norwegian Meteorological Institute, 2012). The yearly average relative humidity of the
outdoor air is close to 80 % (IMO, 2013). After cooling the meal is grinded for even
grain size in a hammer grinder before storage in meal silos or sacks.
2.5Dyno Jet indirect air dryer
Dyno Jet is indirect air dryer which was originally designed as a two -step indirect flame
dryer, where oil is burned in excess of air used for heating up drying air, in an air to air
pipe heat exchanger.
Figure 11: Dyno
jet air drying system (Stord, 1992).
First step: where the meal cake enters the dryer and flows together parallel through the first
step. Volume of first drying-step in the Dyno Jet in Vopnafjordur is around 16 m
(Sigmundsson, 2013) and maximum air flow rate of the drying air, according to
-
7/25/2019 Fish Meal Project
34/87
18
recommendations is an ideal flow around 11 m/s. (Arason, 2001b) When entering the
dryer the meal cake mixes up with drying air resulting in fast flash drying, where majority
of the moisture is removed in a few seconds (7-14 s). The meal cake enters the Dyno Jet at
90 C and the temperature of drying air varies from 220 C to 400 C (Arason, 2001a,
Information from control system, 2012, Heggernes, 1991). Drying air passes through the
first step in less than one second with airspeed approximately 4 m/s, while the journey for
the meal particles takes a bit more time.
Although the temperature of the drying air is high, the wet bulb temperature is only around
50 -60 C (Cengel & Boles, 2007).
2.5.1 Electric heating equipment
With the new electric heating equipment (EHE) the air passes through before entering the
dryer. The electrical heating equipment consists of two main parts, the ribbed steam to air
heat exchanger and the electrical heating elements. The ribbed steam heat exchanger is in
front of the electrical heating elements, the main reason for that alignment is due to a great
pressure drop over the ribbed steam element relative to the electrical elements. The air
stream levels out when passing through the ribbed steam element, to get less distribution in
air speeds inside the electrical heating equipment (Plsson, 2011). Another reason for
positioning the ribbed steam element is for the corrosion and water resistance ability over
the electrical elements, where the drying air coming from the scrubber tower is saturated
with sea water with a relative humidity of 100 %.
-
7/25/2019 Fish Meal Project
35/87
19
Figure 12: Installation of the electrical heating equipment in Vopnafjordur.
Figure 13: DynoJet air drying drum (Stord, 1992).
-
7/25/2019 Fish Meal Project
36/87
20
2.5.2 Air flow in Dyno-Jet and air dryers
The air flow is controlled by closure equipment, which controls the pressure drop in the
Dyno-Jet drying system. The pressure drop is proportional to air speed squared (Crowe,Elger, & Roberson, 2005), so the air flow is sensitive to pressure drop but the control
closure creates only part of the total pressure drop. The air blower is driven on constant
frequency. Prior the air blower the air pressure is lower than atmospheric pressure which
the air blower raises. Adjusting the air flow Dyno-Jet producer recommends that the zero
(pressure equals atmospheric pressure) pressure of the drying system should be between
the pre-dryer and the second dryer. Thereby pressure in pre-dryer is higher than
atmospheric pressure and vice versa in the second dryer. However the zero point transfersby the opening of the closure equipment.
When the drying air leaves the scrubbing tower the relative humidity is 100 % which
means that the air is saturated of water (0.015 kg water / kg air). After passing through the
electric heater the relative humidity drops down below 1 % (Mollier, 2013). After the
drying air has taken up moisture from the fish meal the water content of the air is between
0.260.29 kg water/kg air, which demonstrates that 1 kg of air can absorb 0.245 0.275 kg
water/kg air, depending on temperatures and air flow, and water content of the meal. To
calculate essential air flow for Dyno-Jet 4.5 the total evaporation per hour is divided by the
water content that the air can carry:
Another type of indirect air dryer is Hetland, the main difference between the drying
methods that Hetland has one drying drum which revolves around its central axis. Inside
the drying drum are shovels spreading the meal within the drum and the air flow carries the
meal through the drying drum.
4,500
0.245 0.275
= 16,600 18,400
-
7/25/2019 Fish Meal Project
37/87
21
Figure 14: Hetland drying drum (Arason, 2009).
The Hetland drying system is based on the old flame dryers which where dominant before
steam and indirect air drying.
The main difference between the flame dryer and the indirect air dryers is that the flame is
burned in the same compartment as the meal is dried.
Fire hazard is significantly lower for the indirect air dryers, due to the open flame is in the
same compartment as the fish meal (Karlsson, 2010). In case of electrical brake down the
flow of material stops but it takes the flame longer time to stop, in that case the meal
closest to the flame can be burned with corresponding damage and hazard to the
production. The greatest advantage of the flame dryer over the indirect dryers is the
capacity due to higher temperature at the material inlet, up to 600 C compared to 220 C
to 400 C for Dyno-Jet and Hetland.
-
7/25/2019 Fish Meal Project
38/87
22
Table 1: Capacity of air dryers (Arason, 2013; SR-Mjl, 2013).
Type of dryer
Capacity (kg
vapour/ m h)
Air temperature
(C)
X water content of drying
air (g water / kg air)
Flame dryer 100 900 0.15 - 0.22
Dyno-Jet 30 - 40 220 - 400 0.26 - 0.29
Hetland 25 300 - 400 0.26 - 0.29
Steam dryers 8 - 12* 100 1 - 2
*is in kg vapour / m h
2.5.3 Steam drying Air drying
Drying fish meal in two steps like in HB Grandi fish meal factory in Vopnafjordur,
where steam drying is used as a predryer and air drying as secondarydrying. This new
approach to combine steam prior to air drying, utilizes the advantages of each method. The
steam dryer works well on wet material but the material temperature will rise constantly to
reduce water content, due to decreased heat conduction in dry matter. On the other hand,
in air dryers meal temperature does not rise as much, mainly because the drying time is
dramatically shorter, according toTable 2.
Table 2: Comparison between drying time and meal temperature (Gunnarsson, 2003).
Steam dryer Hetland air dryer Dyno-Jet air dryer
Drying time (min) 80 24 12
Meal temperature (C) 98 78 78
-
7/25/2019 Fish Meal Project
39/87
23
Figure 15: Drying efficiency of steam dryer in terms of water content of fish meal (Arason, 2013).
As can be seen in Figure 15 the drying efficiency begins to decrease when the water
content drops below 38 40 % (Arason, 2013). The main reason for the decreased drying
efficiency is that thermal conductivity of the meal cake is proportional to water content
(Fennema, 1996). When the thermal flux reduces the temperature in the meal closest to the
heating surface increases and the meal can be burned in some cases. When the mealtemperature rises the digestibility of the fish meal decreases (Arason, 2013). This is the
main reason for drying the meal to a water content of about 40 % in the steam dryer and
finish the drying in an air dryer.
0
2
4
6
8
10
12
14
0 15 30 45 60
kgevaporated/(m
h)
Water content (%)
Drying efficiency of steam dryer
-
7/25/2019 Fish Meal Project
40/87
24
2.6Energy balance
In Vopnafjordur the energy comes into the process in two ways, (1) an electric boiler
which, supplies the steam dryer, the cooker and STAHE (steam to air heat exchanger) in
the EHE, (2) and electricity for heating elements in the electric heater. The factory inVopnafjordur is also equipped with oil boiler in case of electrical rationing. The greatest
advantage of the electrical boiler over the oil is the energy efficiency, which is around 98
% for the electrical boiler while for the oil boiler the efficiency is from 78 90 %
(Blndal, 2013). The steam made in the boiler is either circulated or the condensate is
transferred to a storage tank and used as a first rinse in cleansing. The reason why the
condensate is feasible for cleansing is that sodium hydroxide (NaOH) is put into the water
in the boiler to increase conductivity of the water.
Figure 16: Schematic view of energy balance in the fish meal process.
The black line represents the material flow, red line shows the path for waste heat from steam dryer,
red dashed line shows steam from electric boiler and blue dashed line shows condensed steam
circulated to boiler.
-
7/25/2019 Fish Meal Project
41/87
25
The vapour from the steam dryer is sent to first effect in WHE and continues to the second
effect before going to the pre-heater.
2.7Material and energy balances
2.7.1 Mass balance
The fish meal process is a continuous process where inputs and outputs flow continuously
throughout the duration of the process. Inputs are the raw material of fish meal and oil, and
outputs are fish meal, oil and vapour. The process is operated at steady state where the
process variables (all temperatures, pressures, flow rates) do not change with time, except
minor fluctuations when adjusting final moisture in the fish meal (Felder & Rousseau,2005).
Material balance on continuous steady state process on nonreactive material is as follows:
Input = Output + Losses
2.7.2 Energy balance
Drying is the most energy intensive part of the process. It is hard to say how high aproportion of the energy goes to drying as it depends on the raw material, drying
equipment and utilizing of waste heat. If the fish meal plant uses a three step evaporator
and air dryer around 70 % of the energy goes into drying (Arason, 20092010; Nygaard,
2010).
-
7/25/2019 Fish Meal Project
42/87
-
7/25/2019 Fish Meal Project
43/87
27
3 Materials and methodsField experiments were carried out in HB Grandi fish meal factory in Vopnafjordur. These
experiments were conducted over a long period of time because of the varying fishing
season between species, in addition to have an opportunity to measure the old oil heating
equipment.
3.1Mass flow modelling
For the modelling of the mass flow a sampling from the process was needed. In each
sampling, around 500 ml of sample for chemical analysis was taken at each sampling point
in the process shown in Figure 17. After the sampling the samples were cooled down
before freezing and were kept frozen until the chemical analyses were performed.
Table 3: Dates of sampling.
Raw material Date Year
Capelin 2 - 3 March 2011
Herring and mackerel cut offs 14 July 2011Mackerel cut offs 4 August 2011
Capelin 34 February 2012
-
7/25/2019 Fish Meal Project
44/87
28
Figure 17: Measurement points of HB Grandi fish meal factory in Vopnafjordur.
Mass flow model was made for four different raw materials at various seasons and samples
were taken according to Table 3. For estimation of mass flow measuring points were
positioned where the mass flow changes, as can be seen inFigure 17.
The mass flow calculation is based on FFDM (fat free dry matter), and calculated from 100
kg of raw material basis.
-
7/25/2019 Fish Meal Project
45/87
29
Equation for the dehydrating processes with two outputs, the ratio of vapour (Rvapour) and
material (Routput). The ratio of material flow in dehydrating process is calculated as shown
in Equation (1).
(1)
The ratio between vapour and input of dehydrating is described by Equation (2)
(2)
Separation processes with two phase output are calculated according to Equation (3), were
phases one and two from the unit process, such as for example pressing were phase one is
the press liquor and phase two is the press cake.
(3)
The ratio of output 2 is calculated as shown in Equation (4).
(4)The total mass output from process is calculated with Equation (5).
(5)
Each equation can be applied to relevant (dehydrating or seperation) sub processes in the
whole fish meal process (Arason, 2001a).
=
=
1 = , , 2
, 1 , 2
2 = 1 1
=
-
7/25/2019 Fish Meal Project
46/87
30
The samples where sent for chemical analysis to Matis in Nordfjordur. The following
chemical analyses where carried out for calculations of mass flow and to get an overview
of the state of the raw material.
Protein. AE 3. The sample is digested in sulphuric acid in the presence of CuSO4 as a
catalyst. There after the sample is placed in a distillation unit, 2400 Kjeltec Auto Sampler
System. The acid solution is made alkaline by use of a NaOH solution. The ammonia is
distilled into boric acid which is then titrated with H2SO4. The nitrogen content is
multiplied by factor 6.25 to obtain % crude protein (ISO 5983, 2005).
Water content. AE 4. The sample is heated in an oven at 103C+/-2 C for four hours.
Percentage of moisture corresponds to the weight loss(ISO 6496, 1999).
Fat. AE 1. The sample is extracted with petroleum ether, boiling range 40-60C. The
extraction apparatus is 2050 Soxtec Avanti Automatic System (AOCS, 1997).
Salt (NaCl). AE 2. Soluble chloride is extracted from the sample with water. Upon
addition of nitric acid, the solution is titrated with silver nitrate and the end point
determined potentiometrically (AOAC, 2000).
TVB-N. AE6.TVN-B is determined through direct distillation into boric acid. The acid is
titrated with H2SO4(AOAC, 2000).
-
7/25/2019 Fish Meal Project
47/87
31
3.2Air flow measurements for Dyno-Jet
Air flow through the electric Dyno-Jet drying system was measured along with the
pressure drop over the oil heating heat exchangers before the capelin season in the spring
of 2011. The air flow was determined with air speed measurement in the air duct midwaybetween the air blower and the electric heater. The speed of the air in (m/s) was multiplied
with the section area of the air duct (m) to get the air flow in (m/h). Pitot air speed meter
PCE-PFM 2 (PCE - Deutschland GmbH, 2013) owned by Hedinn was utilized for
measuring pressure difference in the air flow and values for temperature and opening of the
control closure for the air flow were extracted from the control system of the factory for
calculations of true air speed.
Single measurements were carried out for each value of the opening of the control closure,
from 30 % - 50 % on 2 % interval, where 50 % is maximum opening of control closure.
The reason for measuring the air flow is to estimate the difference between the air flow in
the electric air heater and the old oil heater. The amount of air flow has a great impact on
the drying system, whether the capacity of the dryer or changing the air speed in various
components of the drying system for example the cyclones and the scrubber tower.
-
7/25/2019 Fish Meal Project
48/87
32
3.3Pressure drop measurement
The pressure drop measurement for the oil heating heat exchanger was carried out for
different air flow, controlled by opening of closure. The pressure drop was measured as
height difference of water in hose connected at the inlet and the outlet of the heatexchanger as shown inFigure 18.
Figure 18: Oil heating heat exchanger and pressure drop measuring hose connected at inlet and outlet
of the heat exchanger. Hose connection are marked on the figure, red ring the air outlet of the heat
exchanger, yellow ring is the measuring position and blue ring is the air inlet to the heat exchanger.
3.4Energy calculations and comparison
The amount of energy used for the fish meal and oil production are taken out from the
operation and registration system InTouch WindowViewer for the factory, the data are
exported to Microsoft Excel where the calculations were carried out.
The electrical energy utilized in the factory is purchased on different bases. Firstly is the
general electricity which is used for driving motors such as screw conveyors, pumps,
centrifuges and compressors for vapour recompression in MVR. Second the insecure
electricity which is only suitable where a backup of another energy source is available, the
-
7/25/2019 Fish Meal Project
49/87
33
insecure electricity is used for the electrical boiler and the electrical heating elements for
the Dyno-Jet dryer. For both the boiler and the heating of the air for Dyno-Jet there is older
equipment powered by oil, for backup in case of restriction of electricity to the factory.
The amount of electricity utilized for each ton of raw material was calculated from the data
of the control system and compared to public data from The Environment agency of
Iceland Green accountingabout oil usage before the electrification, and for comparison
between factories.
-
7/25/2019 Fish Meal Project
50/87
-
7/25/2019 Fish Meal Project
51/87
35
4 Results
4.1Mass and energy flow model for fish mealprocess
The mass flow model was built for four kinds of raw material while the energy
calculations are based on the February capelin (FC) processing. What deviates the FC
processing from the other raw materials was that the factory was run on similar capacity
for five days in a row with an average fish meal production of 116 tons per 24 h.
Calculated to raw material on fat free dry matter (FFDM) basis is around 555 tons per 24 h.
For the other three raw materials March capelin (MC), herring and mackerel cut-offs
(HMCO) and mackerel cut-offs (MCO) the production was not so steady beetween days
and therefore the accuracy of the data and calculation was not the same as for the FC. The
basis for the mass flow model are the chemical analyses carried out in Matis as shown in
Table 4, average of measurements series for each raw material was used for the
calculation. For calculation of the raw material entering the factory the data from the meal
scale where the output of the factory was mesured on a certified scale, and the fish meal
utilization factor calculated from the chemical analysis from Matis.
The following tables and figures apply to Sections4.1.1, 4.1.2,4.1.3 and4.1.4.
-
7/25/2019 Fish Meal Project
52/87
36
Table 4: Results of chemical analysis performed by Matis. Abbreviations: MC March capelin,
HMCO Herring and mackerel cut offs, MCO Mackerel cut offs, FC February capelin, M1
Measurement 1. , measuring points in first column are according toFigure 17.
* For MC-M1 was measured in the process with refractometer.
MC-M1 MC-M 2 MC-M 3 HMCO-M 1 HMCO-M 2 MCO-M 1 FC-M 1 FC-M 2
Water (%) 79.4 77.4 78.1 70.1 72.4 64.0 72.7 73.2
Protein (%) 13.8 14.4 14.1 12.4 11.9 13.3 14.3 13.4Salt (%) 0.6 0.7 0.8 0.8 0.8 0.7 0.8 0.4
Fat (%) 4.3 5.2 5.2 15.4 12.6 19.5 10.5 10.4
FFDM (%) 16.3 17.4 16.7 14.5 15.0 16.5 16.8 16.4
Water (%) 93.9 94.2 72.1 69.9
Fat (%) 2.1 1.3 11.7 14.9
FFDM (%) 4.0 4.5 16.2 15.2
Water (%) 91.3 94.2 86.1 73.9 75.4 77.5 79.6 78.6
Fat (%) 1.7 1.0 7.9 19.2 16.5 15.5 13.8 12.6
FFDM (%) 7.0 4.8 6.9 8.1 7.0 6.6 8.8
Water (%) 51.6 51.7 49.9 45.5 46.2 43.2 51.5 53.2
Fat (%) 4.4 4.5 3.7 4.9 6.1 2.2 4.6 4.3
FFDM (%) 44.0 43.8 49.6 47.7 54.6 43.9 42.5
Water (%) 92.7 93.2 92.4 81.5 80.5 77.9 81.0 80.9Fat (%) 3.0 2.6 3.5 13.1 13.7 15.8 13.6 12.4
FFDM (%) 4.3 4.2 4.1 5.4 5.8 6.3 5.4 6.7
Water (%) 59.8 61.8 61.2 63.2 66.9 63.5 70.2
Fat (%) 4.6 3.8 5.1 4.2 6.3 3.8 4.0
FFDM (%) 35.6 34.4 33.7 32.6 26.8 32.7 25.8
Water (%) 95.5 95.3 95.3 93.3 93.1 93.3 93.1 90.8
Fat (%) 0.3 0.4 0.4 0.7 0.6 0.7 0.9 1.4
FFDM (%) 4.2 4.3 4.3 6.0 6.3 6.0 6.0 7.8
Water (%) 73.9 88.7 89.0 90.4 88.1 86.7
Fat (%) 2.2 1.1 3.7 1.2 1.5 1.8
FFDM (%) 23.9 10.2 7.3 8.4 10.4 11.5
Water (%) 67.1 69.2 73.0 79.1 75.1
Fat (%) 2.9 5.2 3.1 2.5 3.5FFDM (%) 30.0 25.6 23.9 18.4 21.4
Water (%) 65.2 58.4 66.8 61.9 59.3 58.9 61.8 65.2
Fat (%) 2.1 3.8 2.3 4.3 6.9 4.6 3.8 4.2
FFDM (%) 32.7 37.8 30.9 33.8 33.8 36.5 34.4 30.6
Water (%) 35.8 26.9 29.1 31.3 48.6 43.1
Fat (%) 5.5 8.0 7.4 6.0 7.6
FFDM (%) 58.7 62.9 61.3 45.4 49.3
13 Dyno Jet dried Water (%) 7.7 7.3 7.8 8.1 9.4
Water (%) 7.3 6.4 7.6 8.0 7.6 9.3
Protein (%) 69.0 66.5 66.7 70.2 69.5
Salt (%) 5.4 6.7 0.1 0.2
Fat (%) 8.0 8.2 9.6 11.2 12.4
FFDM (%) 84.7 85.4 82.4 81.3 78.314 After cooling and mill
10 Concentrate MVR + WHE
11 Prior drying
12 Steam dried
7 Decanter sludge
8 Stickwater
9 Concentrate MVR *
5 Presscake
6 Separated press l iquor
1 Raw material
3 Pre filter
4 Press liquor
-
7/25/2019 Fish Meal Project
53/87
37Figure 19: Relative mass flow for all for raw materials, MC* in vapour WHE II is the evaporation for both MVR and WHE II run in parallel.
Fish
meal
Fish
oil
MCO 100
HMCO 100
FC 100
MC 100
Raw material
MCO 20.0
HMCO 17.6
FC 25.1
MC 28.8
Presscake
MCO 80.0
HMCO 82.4
FC 74.9
MC 71.2
Press liquor
MCO 77.3
HMCO 76.7
FC 69.6
MC 67.4
Separated press liquor
MCO 12.2
HMCO 10.3
FC 9.0
MC 1.9
Fish oil
MCO 65.1
HMCO 66.4
FC 60.6
MC 65.5
Stick-water
MCO 18.6
HMCO 19.7
FC 22.4
MC -
Vapour MVR
MCO 30.2
HMCO 32.0
FC 17.2
MC* 55.5
Vapour WHE II
MCO 16.3
HMCO 14.7
FC 21.0
MC 10.0
Concentrate
MCO 39.0
HMCO 38.0
FC 51.4
MC 42.6
Prior drying
MCO 15.8
HMCO 17.6
FC 16.1
MC 17.0
Vapour SD
MCO 23.2
HMCO 20.4
FC 35.3
MC 25.6
Steam dryedMCO 6.0
HMCO 5.2
FC 14.3
MC 7.8
Vapour D-J
MCO 17.3
HMCO 15.3
FC 20.9
MC 17.7
Meal
MCO 2.7
HMCO 5.7
FC 5.3
MC 3.7
Decanter sludge
-
7/25/2019 Fish Meal Project
54/87
38
Figure 20: Results of chemical analysis for February capelin set up in a process diagram.
Fish
meal
Fish
oil
Water (%) 73.0
Fat (%) 10.5
FFDM(%) 16.6
-Protein (%) 13.9
February capelin
Raw material
Water (%) 52.4
Fat (%) 4.5
FFDM (%) 43.2
Presscake
Water (%) 66.9
Fat (%) 3.9
FFDM (%) 29.3
Decanter sludge
Water (%) 79.1
Fat (%) 13.2
FFDM (%) 7.7
Press liquor
Water (%) 81.0
Fat (%) 13.0
FFDM (%) 6.1
Separated press liquor
Water (%) 92.0
Fat (%) 1.2
FFDM (%) 6.9
Stick-water
Water (%) 77.1
Fat (%) 3.0
FFDM (%) 19.9
Concentrate
Water (%) 63.5
Fat (%) 4.0
FFDM (%) 32.5
Prior drying
Water (%) 45.9
Fat (%) 6.8
FFDM (%) 47.4
Steam dryedWater (%) 8.4
Fat (%) 11.8
FFDM(%) 79.8
-Protein (%) 69.8
Meal
-
7/25/2019 Fish Meal Project
55/87
39
4.1.1 March capelin whole and cut offs
When March capelin (MC) measurements were performed the mechanical vapour
recompression (MVR) and waste heat evaporator (WHE II) where run in parallel but not in
series such as in the following measurements. As can been seen in Figure 19 the mealutilization was 17.7 % and fish oil utilization is 1.9 %. As stated inTable 7,2/3 of drying
evaporation is carried out in the steam dryer and nearly 70 % of the total evaporation is
carried out in the MVR + WHE, but the dry matter content measurement was done with
refractometer in the factory and is not as accurate as other chemical analysis. When the
factory is not run on full capacity a larger share of the drying is needed in the steam dryer,
so the WHE II can get enough vapour for the evaporation. This will result in excessively
dry meal from the steam dryer following decreased digestibility of the meal if dried below38% water content in a steam dryer (Arason, 2011). As shown inTable 4 water content in
the partly dryed meal from the steam dryer is 35.5 % which is below the recomended water
content for the best results in terms of digestibility. InTable 7 it can be seen less than tenth
of the total evaporation is carried out in the Dyno-Jet air dryer, which is far below the
capacity of the Dyno-Jet.
4.1.2 A mix of herring and mackerel cut-offs
In the summer the pelagic vessels are fishing herring and mackerel and their aim is to bring
to shore highest quality raw material for processing and freezing in HB Grandi pelagic
processing plant in Vopnafjordur. When the fishing for human consumption, the raw
material for fish meal and fish oil processing are the cut-offs and fish that does not meet
quality requirements. The quantity of raw material for fish meal and fish oil processing is
determined by the capacity of the processing plant and the quality of the fish. When this
measurement was carried out the fish meal production was close to 100 tons of meal over
the 24 hour period, which can be calculated to 620 tons of raw material with the results of
material balance which states 15.3 % fish meal utilization of the RM. The flow meter
before the presses showed values between 625 640 m/day which is close to the
calculated flow of RM, showing quite good accuracy of the flow meter. The reason for
determining the capacity of the factory from data out of the meal scale rather than the flow
meter, is the lack of accuracy of flow measurements of inhomogeneous material prior to
the presses.
-
7/25/2019 Fish Meal Project
56/87
40
To estimate the total evaporation in each stage of the water removal in the process it is
necessary to use the ratio figures fromTable 7 and multiply them by the total input of raw
material. The first thing to notice is the amount of evaporation by the steam dryer
compared to the Dyno-Jet, 4.55 mt/h compared to 1.34 mt/h in the Dyno-Jet. The main
reason for removing such a great share of the water in the steam dryer is to have enough
waste heat to drive the WHE II, which evaporates 8.27 mt/h from the 4.55 mt/h of vapour
from the steam dryer resulting in 0.55 kg vapour / kg evaporated (more details are inTable
9).
4.1.3 Mackerel cut-offs
For the mackerel cut-offs only one sampling was made because of a breakdown of a pump
in the stick-water section, and therefore the production was stopped immediately. Due to
the fact that only one sampling is the base for mass flow calculation the accuracy of the
input to the model was not as great as for the other raw materials, the variation between
samplings can be seen inTable 4. The flow meter before the presses showed 398 m/day
which is far from the calculated flow of raw material. According to the meal scale the total
meal production was around 37 mt of fish meal over the 24 h period resulting in a raw
material flow of 214 mt/day. The main reason for this difference is some fluctuations
because of the immediate stop of production.
As pointed out previously in Section3.1 to estimate the total evaporation in each stage of
the water removal in the process figures from Table 7 and Table 8 are needed for the
calculation. The same situation is here as for the mix of herring and mackerel, the amount
of evaporation by the steam dryer compared to the Dyno-Jet is about the triple of the D-J
(seeTable 9). The evaporation in the steam dryer is 1.41 mt/h compared to 0.54 mt/h in the
Dyno-Jet. The same reason as before is for the removal of water in the steam dryer, to have
enough waste heat for the WHE II, which evaporates 2.69 mt/h from the 1.41 mt/h of
vapour from the steam dryer resulting in 0.52 kg vapour / kg evaporated. The utilization of
the WHE II is very good compared to the design criterion based on empirical prerequisite
around 0.67 kg vapour / kg evaporated (Arason, 2009).
-
7/25/2019 Fish Meal Project
57/87
41
4.1.4 February capelin whole
The production of fish meal was on average 116 mt over a 24 h period when the sampling
took place, using the fish meal utilization 20.9 % the amount of raw material can be
calculated to 555 mt / day. The reason for making more detailed analysis on the Februarycapelin (FC) measurements is because of stable fish meal production the days before and
after.
First of all is to start with an estimation of the total evaporation in each stage of the water
removal in the process. Compared to the measurements for the HMCO and MCO the
situation here is the vice versa, because of better knowledge of the control of drying in
terms of digestibility of the meal. Therefore the amount of evaporation by the steam dryer
compared to the Dyno-Jet is nearly the same as for the D-J (seeTable 9). The evaporation
in the steam dryer is 3.72 mt/h compared to 3.31 mt/h in the D-J. The same reason as
before is for the removal of water in the steam dryer, to have enough waste heat for the
WHE II, which evaporates 3.98 mt/h from the 1.41 mt/h of vapour from the steam dryer
resulting in 0.94 kg vapour / kg evaporated. The utilization of the WHE II is low compared
to the measurements of the HMCO and MCO. The reason for that difference is found in
the difference in physical properties of the stick-water as stated in Section 2.4.1,also the
temperature in the storage tanks for the RM plays a role and the degradation differencebetween cut-offs and whole fish. All those factors stated here above have the effect of
making it harder to press the liquid phase out from the muscles of the capelin. For that
reason more energy is needed for cooking the capelin and therefore less energy can be
utilized from the waste heat coming from the steam dryer.
Table 5 andTable 6 show the state of the factory and demonstrate the variations of values.
-
7/25/2019 Fish Meal Project
58/87
42
Table 5: State of factory when measurement 1 was carried out.
Measurement 1 - February capelin
Time
Flow
to
press(m/h)
Meal
scale(kg/h)
In-let
temperatureDyno-Jet (C)
Number
of electric
elements(n)
Electrical
usage of
elements(kW)
Steam
element
pressure(bar)
Opening
of air
flow
closure(%)
Estimated
air flow(m/h)*
Pressure
drop over
EHE(mmH2O)
Raw
material
storage('000 kg)
Blood
water
storage
('000kg)
Flow
blood
water(m/h)
Steam
dryer
pressure(bar)
Steam boiler
power
consumption(kW)
Total power
consumption(kW)
12:13 30.6 4987
12:15 30.3 4987
12:16 29.3 291 7 1400 8.4 50 42,484 29 339 40.6 4.4 8300 9700
12:30 28.0 4787
12:36 29.2 4786 293 7 1400 9 50 42,484 30 331 39.9 4.5 4.4 8300 9700
12:40 30.5 4975
13:00 29.6 4783
13:40 30.2 4787
13:42 30.8 4787 310 8 1600 8.4 50 42,484 29 302 37 4.4 8500 10100
14:50 29.9 4608 315 8 1600 8.2 55 47,968 29 273 34.1 4.5 4.4 8300 9900
15:10 29.2 4417
15:30 30.0 4428
15:50 29.6 4242
16:10 28.7 4261
17:52 29.7 2705 311 8 1600 7.5 46 39,915 29 199 25.9 0.1 8300 9900
18:28 29.7 7952 308 8 1600 8.0 45 39,377 30 183 24.1 4.6 8100 9700
Average 29.7 4766 305 8 1533 8.3 49 42,452 29 271 34 4 4.4 8300 9833
*Estimated air flow is according to air flow measurements described in Section4.4.
-
7/25/2019 Fish Meal Project
59/87
43
Table 6: State of factory when measurement 2 was carried out.
Measurement 2 - February capelin
Time
Flow
to
press
(m/h)
Meal
scale
(kg/h)
In-let
temperature
Dyno-Jet (C)
Number
of
electric
elements
(n)
Electrical
usage of
elements
(kW)
Steam
element
pressure
(bar)
Opening
of air
flow
closure
(%)
Estimated
air flow
(m/h)*
Pressure
drop over
EHE
(mmH2O)
Raw
material
storage
('000
kg)
Blood
water
storage
('000
kg)
Flow
blood
water
(m/h)
Steam
dryer
pressure
(bar)
Steam boiler
power
consumption
(kW)
Total power
consumption
(kW)
7:13 30.4 5588
7:16 29.9 5392 272 6 1200 8.1 50 42,484 32 1427 6.3 0.4 8800 10000
7:23 29.0 5196
7:37 30.5 4812 277 6 1200 8.1 50 42,484 32 1415 6.6 2.9 8700 9900
7:40 28.2 4812
8:25 29.6 5006
9:56 28.7 5196 299 7 1400 8.5 50 42,484 31 1337 21 3.1 8500 9900
10:08 30.8 5194
Average 29.6 5149.5 282.7 6.3 1266 8.2 50 42,4843 31.7 1393.0 11.3 2.1 8666 9933
*Estimated air flow is according to air flow measurements described in Section4.4.
-
7/25/2019 Fish Meal Project
60/87
44
Out of the information from the plant control system and given that steam to air heat
exchanger is running on 2000 kW an estimation of the energy usage in the factory shows
that 34 % of the total energy usage goes to the Dyno-Jet air dryer in the drying of the
February capelin.
Table 7: Evaporation for all raw materials.
Table 7 shows the difference between total evaporation of the four kinds of raw material
(RM), the column in front in each RM shows the percentage evaporated of the RM and the
rest is fish meal and fish oil. The utilization of fish meal and oil for all the raw materials
can be seen in Figure 19.The most interesting part is to compare the evaporation ratiobetween the evaporators (MVR and WHE II) and the dryers (Steam and D-J). First of all is
the difference in ratio of the Dyno-Jet for the February capelin (FC) and the other, as stated
before when the production of FC the factory was run close to its maximum capacity so
increased performance was needed in the drying. For the FC the fifth of the evaporation
was carried out in the D-J which is double than for the other RM.
Equipment
Percentage
of RM
evaporated
Percentage of
the total
evaporation
Percentage
of RM
evaporated
Percentage of
the total
evaporation
Percentage
of RM
evaporated
Percentage
of the total
evaporation
Percentage
of RM
evaporated
Percentage
of the total
evaporation
MVR 19.7% 26.4% 18.6% 26.3% 22.4% 32.0%
WHE II 55.5% 69.1% 32.0% 43.0% 30.2% 42.8% 17.2% 24.6%
Steam dryer 17.0% 21.2% 17.6% 23.6% 15.8% 22.4% 16.1% 23.0%
Dyno - Jet 7.8% 9.7% 5.2% 7.0% 6.0% 8.5% 14.3% 20.4%
Total 80.3% 100.0% 74.5% 100.0% 70.6% 100.0% 70.0% 100.0%
FCMC * MCOHMCO
-
7/25/2019 Fish Meal Project
61/87
45
4.2Energy balance for evaporation and drying
4.2.1 Comparison of waste heat evaporator capacity betweenraw materials
Table 8: Waste heat evaporation in terms of waste heat input from steam dryer for different rawmaterials.
Raw material
Steam dryer
evaporation (kg/100 kg
RM)
WHE II evaporation
(kg/100 kg RM)
WHE II efficiency (kg
steam / kg
evaporated)
Mach capelin 17.0 55.5* MVR + WHE II
Mackerel cut-offs 15.8 30.2 0.523
Herring and mackerel cut-offs 17.6 32.0 0.550
February capelin 16.1 17.2 0.936
As can be seen in Table 8 the evaporation in WHE II for February capelin is just above
half of the evaporation for mackerel and herring cut-offs. There are two main reasons for
the low evaporation for the capelin, first the viscosity of capelin stick-water is significantly
higher (see Figure 9). Because of high water holding of capelin the pressing is more
dependent on adequate cooking, so a larger part of steam dryer waste energy is sent
through the WHE II to the pre heater.
Table 9: Evaporation in water removal processes.
HMCO MCO FC
Equipment
Total
evaporation
(000 kg / h)
Total
evaporation
(000 kg / h)
Total
evaporation
(000 kg / h)
MVR 5.09 1.66 5.18
WHE II 8.27 2.69 3.98
Steam dryer 4.55 1.41 3.72
Dyno-Jet 1.34 0.54 3.31
Total 19.25 6.30 16.19
Table 10: Comparison between processing properties mass flow of raw materials.
Raw material
Press cake
ratio (%)
Press
liquor
FFDM (%)
Stick-
water
FFDM (%)
Final
concentrate
FFDM (%)
March Capelin 28.8 5.9 4.3 -
Mackerel cut-offs 20.0 7.0 6.0 23.9
Herring and mackerel cut-offs 17.6 7.5 6.2 27.8
February capelin 25.1 7.7 6.9 19.9
-
7/25/2019 Fish Meal Project
62/87
46
4.3Energy utilized for processing fish meal andoil.
The only public data on energy usage of the Icelandic fish meal factories are available in
green accounting that the factories must hand in to the Environment agency of Iceland. Thecomparison inTable 11 shows very variant total energy usage between factories, attributed
to many factors. First of all is the equipment, especially dryers and evaporators, which is
different from one factory to another, utilization of waste heat in the process and total
processing days of the year. Another element that is missing in the equation is that some of
the factories like HELSVN, SEY-SVN and AKNHBG are mostly processing capelin
meal while the other are processing fish meal from all of the pelagic species caught around
Iceland. As stated in Section4.1 the total evaporation varies between the pelagic speciesand in addition to the state of the raw material, whether it is cut-off or whole. Those facts
can illustrate partially the difference in total energy usage in the last row ofTable 11,while
another part of the difference can be attributed to differences in factory setup (shown in
Table 12).
First of all is the lowered total energy usage for Vopnafjordur from 2010 to 2011 where the
main change was the electrification of the DynoJet (D-J) air dryer, before that change the
factory was run by the electrical boiler and the D-J was powered by oil. For over the year
2010 the energy used by the D-J was close to one third of the total energy usage compared
to over one third for whole capelin in February 2012. Because when the D-J was driven by
oil the focus was to take out the largest share of the moisture in the steam dryer, with
regard to minimized energy cost although the digestibility of the meal would be lowered.
After the implementation of the EHE to the D-J the impact of the energy cost factor has
been lowered, the focus has been to have at least 40 % water content in the meal when
leaving the steam dryer. Also when the factory is close to maximum capacity larger shareof the moisture must be taken out in D-J.
It can be assumed by the data inTable 11 that the saving of energy between the years 2010
and 2011 is around 76 MJ / 000kg RM and associated with only the energy for D-J, which
is 537,75 MJ / 000kg RM a rough calculation can lead to close to a 15 % energy savings
by the EHE. The lowered total energy usage of the EHE compared to the oil heating unit
can be explained in many ways, first of all there is no waste heat sent through the chimney.
-
7/25/2019 Fish Meal Project
63/87
47
In addition the startup time for the EHE very short in comparison with the hours it takes
the stone facing in the furnace to heat up at a low constant heat rate (Sigmundsson, 2013).
Other factories that is worth to compare to Vopnafjordur due to similar set up is the factory
in Nordfjordur also with a steam dryer as a pre dryer and air dryer as a secondary dryer.
Nordfjordur factory is also equipped with steam boiler and situated close to processing
plant. As shown in Table 11 the total energy usage is dramatically higher in Nordfjordur
than in Vopnafjordur, and that is probably explained by the increased water content in the
raw material. The increased water content is due to their transporting system, where the
raw material is pumped through a 600 m long pipe under the harbor from the pelagic
processing plant. On that way the raw material soaks up water which results in increased
energy usage for the fish meal production.
-
7/25/2019 Fish Meal Project
64/87
48
Table 11: Energy consumption comparison between factories (UST, 2012).
Vopnafjordur -HBG NOR - SVN SEY - SVN HEL - SVN AKN - HBG TOH - IV VES - FES VES - VSV
2011 2010 2011 2011 2011 2011 2011 2011 2011
Raw material
(000kg) 46,901 46,138 92,778 34,827 28,203 27,190 38,882 49,815 57,049
Electricity usage
(MWh) 21,444 15,174 38,467 11,233 9,896 2,215 2,446 4,327 12,739
Oil usage (000kg) - 627 2,334 770 593 1,298 1,810 2,011 1,375
Electricity
efficiency (kWh/
000kg RM) 457.2 328.9 414.6 322.6 350.9 81.5 62.9 86.9 223.3
Oil efficiency (kg/
000kg RM) 0 13.6 25.2 22.1 21.0 47.7 46.5 40.4 24.1
Energy usage of
electricity (MJ/
000kg RM) 1645.99 1183.98 1492.62 1161.18 1263.17 293.31 226.43 312.68 803.90
Energy usage of
oil (MJ/ 000kgRM) 0 537.75 995.66 874.44 832.31 1889.02 1841.81 1371.54 819.02
Total energy
usage (MJ/
000kg RM) 1646 1722 2488 2036 2095 2182 2068 1684 1623
-
7/25/2019 Fish Meal Project
65/87
49
When comparing the fish meal factories inTable 11 for the year 2011, the most interesting
number for comparison is the total energy consumption, as shown inTable 12 the setup of
the factories is different. As stated in Section2.5.3before the main difference between air
dryers and steam dryers in terms of energy usage is for the steam dryer the possibilities of
recycling waste heat are better than for air dryers. Another thing that must be taken in to
account is the energy efficiency of the boilers, where factories have both oil and electrical
boilers they are normally running only on the electrical boiler. Because the difference
between energy utiliaztion of electrical and oil boliers, the electrical bolier has energy
efficiency of 97 98 % against oil boiler can be varying from 78 90 % depending on
condition and age (Blndal, 2013; Fulton, 2013; Geankoplis, 2003). Besides the difference
in setup of the factories is that the raw material they receive is different. The factories can be
split up in to two categories: first are the factories stationed along with pelagic processing
plants, Vopnafjordur, Nordfjordur, orshofn and both of the factories in Vestmannaeyjar.
Other factories are indipendent from processing plants except roe extracting of capelin in
Akranes and Helguvik (Arason, 2013; Jnsson, 2013).
The amount of energy used to run pumps, motors for screw conveyors, ventilation and
lightning in the factories, is taken from electrical energy data onfactories driven by oil
(AKN-TOH and Faskrudsfjordur) (UST, 2012; Arason & Gslason, 2012).
Table 12: Setup of factories inTable 11.
Factory Oil Boiler
Electrical
boiler
Steam
dryer
Dyno -
Jet dryer
Hetland
dryer
Vopnafjordur x x x x
Nordfjordur (NOR) x x x x
Seydisfjordur (SEY) x x x x
Helguvik (HEL) x x x xAkranes (AKN) x x
orshofn (OH) x x
Vestmannaeyjar (VES-FES) x x
Vestmannaeyjar (VES-VSV) x x x
-
7/25/2019 Fish Meal Project
66/87
50
Table 13: Energy cost comparison between heavy fuel oil (HFO) and electricity (Hgnason, 2012;
Ptursson & Bjrgvinsson, 2013).
Energy source Cost Unit
Cost per
energy unit
(ISK/MJ)
Oil (HFO) 110 (ISK/L) 2.60
Electricity 5.15 (ISK/kWh) 1.43
4.4Air flow and pressure drop in heatexchangers
4.4.1 Air flow for the electrically powered Dyno Jet air dryermeasurement
Results of the air flow measurements can be seen inFigure 21 in terms of air flow against
opening of the closure. The opening of the closure controls the pressure drop in the drying
system, and therefore the total amount of air flow. The air flow is almost linear with the
opening of the controlling closure as shown inFigure 21.
A table with more detailed results of the air flow measurements can be found in Appendix
A.
Figure 21: Results of air flow measurements.
24,000
26,000
28,000
30,000
32,000
34,000
36,000
38,000
40,000
42,000
44,000
30 32 34 36 38 40 42 44 46 48 50
Airflow
(m/h)
Closure opening (%)
Air flow vs. closure opening
-
7/25/2019 Fish Meal Project
67/87
51
Measured maximum air flow was greater than expected (Arason, 2009), which is in
harmony with lower pressure drop in the electric air heater than measured for the oil heat
exchanger.
4.4.2 Pressure drop over oil heating equipment
In addition pressure drop over the oil heating heat exchanger was measured for two different
air flows, results can be seen inTable 14.The pressure drop was measured with a hose filled
up with water as stated in Section3.3 and the change of the water level in the hose is half the
pressure drop over the oil heating equipment (Pressure measurement, 2013). Photos from thepressure drop measurements are situated in Appendix B.
The measured pressure drop over the oil heating heat exchanger was 38 mmH2O to 84
mmH2O compared to the maximum press